Tohban Report 2015-07-15
From RHESSI Wiki
|Start Date:||01 Jul 2015|
|End Date:||08 Jul 2015|
|List all reports|
VERY quiet week, with not even an M flare. There are only 3 active regions on the disk. It's expected that it will stay quiet.
Side note: We should soon start getting an occasional image from STEREO-A again since it's come out of conjunction. Since it's gone all the way to the far side of the Sun and has come out of conjunction, STEREO-A will now be showing us the side of the Sun that will rotate onto the visible disk. There's still no word from STEREO-B.
How many GOES flares occurred?
Flares above B, C, M, X class were 11 6 0 0
And how many of these are listed in the RHESSI flare list?
Flares above B, C, M, X class were 6 5 0 0
And how many had EXCELLENT coverage?
Flares above B, C, M, X class were 0 0 0 0
There were RHESSI flares/GOES flares 75 / 17 over the time range 08-Jul-15 15-Jul-15
The SSR is still not emptying. It's only getting down to ~21% at the end of the daily pass set (max 47%, min 18% for the week). The level doesn't seem to be rising, but it's not going down. Source of large fill is unclear. There's been almost NO activity. We're in normal/vigorous decimation. We are taking rear eclipse data but not fronts -- we could kill the rears if we want to clear out the SSR. (Decision during the Ops meeting was to yes, stop nighttime events. Tohban will make the call for this, probably tomorrow.)
Occasional 5- or 10-min gaps in SOH, no gaps in monitor rates.
Detector 3 and 4 had low front livetimes, so the front fast thresholds were changed on both of these detectors on July 9:
2015-191-19:20:47 /IDPUTABLE3 OFFSET=FRONTFASTDAC 2015-191-19:20:58 /IDPLOAD VALUE=0x40 ; was 0x30
2015-191-19:21:17 /IDPUTABLE4 OFFSET=FRONTFASTDAC 2015-191-19:21:25 /IDPLOAD VALUE=0x80 ; was 0x70
This action was effective in decreasing fast rates and increasing livetime in both detectors.
Due to low livetimes, the rear fast thresholds on detectors 1, 6, and 7 were raised on July 13:
2015-194-16:33:58 /IDPUTABLE1 OFFSET=REARFASTDAC 2015-194-16:34:12 /IDPLOAD VALUE=0x50 ; was 0x45
2015-194-16:34:40 /IDPUTABLE6 OFFSET=REARFASTDAC 2015-194-16:34:52 /IDPLOAD VALUE=0x80 ; was 0x70
2015-194-16:35:07 /IDPUTABLE7 OFFSET=REARFASTDAC 2015-194-16:35:15 /IDPLOAD VALUE=0x50 ; was 0x45
These actions were successful in reducing fast rates and freeing up livetime. Action could be repeated, since the LT levels on these detectors (plus D8) are now ~65-80%.
Front status notes:
- D6 has very high resets (essentially flatlined with no more orbital variation) and high event rates. This could be one of the causes of the high SSR fill. But HV is low (<2kV) so maybe dangerous to drop more. This behavior has been consistent for weeks. Investigation of whether D6 is returning good data would be useful.
- D4 and D6 have high slow rates.
- D8 and D9 have very low livetimes. (~80% for D8 and <50% for D9). These have been steadily dropping since about June 24. Before that, there were some slow drops and slow rises. Changes are very gradual, with no large jumps. (Plots of this would be useful.) Both detector fast thresholds were raised on June 11, with only a tiny effect on the livetime. Interesting note: D9’s fast rates dropped dramatically at this change, but D8’s did NOT. Neither detector increased much in LT. This situation persists to this day — D8’s fast rates are quite high. D9 has low slows, fasts, and resets.
- D4's LT is still a little low (90-95%), though the last fast threshold raise helped.. Could raise fast threshold even more.
Rear status notes:
- Livetime is still a bit low on D1, 6, and 7. Might be worth raising the fast thresholds again. Best to check with gamma-ray community since changing the fast threshold affects actual counts collected for the rear segment. Update: Albert recommends raising D6 threshold (a little) only, not the others.
- D8 and D9 have very high reset rates in the rears (but not the fronts). Could be time for a step down in HV. Of course, the D9 rear segment is effectively not working, so may not gain much.
In the last tohban week, we performed an experiment where we had changed the deadtime threshold for shutter insertion from the nominal 5.1% to 10%. This was intended to get better low-energy data for Brian and others to do low-energy HXR investigations. However, we found that with that new deadtime threshold, the shutter inserted not that much later (i.e. when the flare gets bright, it gets really bright, and so you'd have to change the threshold by a LOT to make a difference). Also, in some of the shutter-out intervals the pileup effect was quite high. It didn't seem productive to continue this experiment, so the deadtime threshold was returned to the nominal 5.1%. It was discussed at last week's meeting that at some point in the future we may disable the shutter altogether in order to get Brian his low-energy data for an entire flare, but we will not try this yet.
The return to nominal deadtime threshold for shutter insertion was done on 2015-189-20:05:38 UTC (July 8, 2015).
On July 9 at ~03:25 UTC, the thin attenuator went in and stayed until Jeremy noticed it and commanded it out (in for a total of about 20 hours total). On investigation, it was determined that dead time on 1 and 7 had risen to a level which wouldn't allow the thin shutter to come out. (Typical livetime was 98.5%, which is less than the deinsertion threshold of 98.8%.) No individual problems were found in either D1 or D7 (no high fast or slow rates or resets). Furthermore, D1 and D7 LT were tracking each other well, so it was not an individual detector problem.
It was therefore decided to change the attenuator deinsertion dead time threshold to 1.96% deadtime (98.04% livetime). The deinsertion deadtime threshold change was done on July 10. Jeremy's summary: "The attenuator threshold value was changed to "5" at 2015-191-19:22:12 (Friday July 10, 2015). The Thin attenuator should now come out when the average live time of G1 and G7 is above 98.04%"
There have been NO attenuator motions since, but this is because there haven't been flares!
During today's Ops meeting, we discussed different potential schemes for managing the shutter motions. There's the Mark/Gordon idea of setting the shutter-out threshold at a level where the shutter will stay in, and then using ATS commands to force the shutter once per orbit, just before or just after eclipse. That way the shutter will come in when things start getting bright and will just stay in until we get to the once-per-orbit forcible removal. This would prevent the shutters from "toggling," i.e. coming out for a count check every 4 minutes. It would mean that we lose counts at low energies in the decay phase of the flare, and also for any small flares that happen to occur afterward in that orbit, but we won't have so much toggling.
It was decided that we'd weigh the ramifications of this (in terms of ATS loads, what would happen with the thick shutters, etc) and rediscuss at next week's Ops meeting, making a decision on whether to implement the scheme at that time. Anyone who has an objection or wants to weight in on this scheme should attend or call into next Wednesday's meeting.
An alternate scheme proposed by Martin is to do a forced de-insertion every ~10 minutes or so, then allow the shutter to (automatically) come back in if rates are high.
|Night time data (fronts)||No night events|
|Night time data (rears)||Taking night events|
|Require extra passes?||No|
|Attenuator operation||Once got stuck in A1; see notes above.|